Annie weighs in with an astute observation on the work of SANAA:

Nishizawa used the term 'atmosphere' over and over again. Maybe it was just a coincidence of misaligned translation, but between his use of that term and the inescapable figure of the bubble in every project I couldn't help but think about Peter Sloterdijk's whole the-future-of-architecture-is-foam position.
To extend this observation, the materiality of the recent SANAA, Nishizawa, and Sejima projects enforces a 'connected-isolation' in some subtle ways. When not literally foaming in plan like the Toledo Pavilion or bubbling in elevation such as Zollervein, cells of space are made ambiguous through the use of material and light to dissolve spatial boundaries. Sloterdijk suggests the importance of architecture in the formation of a foam by noting, "no dough without a vessel to form it in; no mass without a hand that knows the purpose of its kneading" (Log 9). In SANAA's architecture we've found a vessel.

House A in Tokyo uses a combination of loose curtains and taut, scrim-like curtains to allow the inhabitants of the house to parcel the space according to their whims. Nishizawa spoke about this in very pragmatic terms: sometimes you go to the bathroom to go, but sometimes you use the room for any number of things that actually don't require as much privacy. In contrast to the bombastic re-building of rooms proposed by Cedric Price's Fun Palace, Nishizawa leaves the rooms intact but uses curtains (and the privacy they afford) to alter the effective definition of a space. This acknowledgment of the life of the rooms in a home actually frees the space that the room contains, as if dislocating it from its socket, and allows for slippage between the spatial and programmatic definition of each room/space in the house. Here, the figure may be mutable but the figure/ground relationship is clear at all times due to a rather small and narrow plan.

Quite the opposite, the Toledo Glass Pavilion is a radical play in composition where the academic fluctuation of figure/ground in plan is mirrored experientially by a constant dissolving and re-constitution of the vertical surfaces as reflections are caught between the multiple layers of glass. Curiously, although this building in its completed state is much more literally transparent than previous SANAA projects, and so employs reflection effects on an almost unprecedented level, the renderings were consistent with other SANAA projects.

The architects consistently choose to represent their world as a whitewashed wonderland of planar ambiguity where light pours in through openings to melt the contents of the model into a single whole. One knows that the thing in front of them is comprised of multiple spaces and planes, that forms are articulated, that difference exists, and yet this is drowned out by the mediating medium of light. So while Sejima describes the New Museum in NYC as "pearly-gray volumes piled with artful carelessness, each off-center to the one below," what she shows is a single volume of ambiguous definition.

The buildings of SANAA are the cap of a bone-dry cappuccino: foam existing without its progenitor. Whereas Koolhaas produces a destabilizing atmosphere by actively opposing his architecture to a known trope, SANAA avoids reference to prototypes. Rather than being destabilized or agitated by a constant variability in the way that we perceive the buildings, this ontological churning becomes the default mode for SANAA's architecture. This is an architecture of vacillation, of multiplicity, of foam.

In Toledo, the New Museum, and House A (as well as much of their other work), Sejima & Nishizawa develop a planning strategy that positions the building as an insulating medium for a foaming, active use and re-use, seeing and re-seeing of the building. At the same time, this academic understanding of the building is bolstered by the development of a rich experience with the careful use of materials to produce ambiguity through effects of light and reflection.

If Sloterdijk claims that an architecture of foam should allow for "both the isolation of individuals, and the concentration of isolated entities into collective ensembles of cooperation and contemplation" (Log 9), we've found the hint of such a condition in the work of SANAA. However, the question of agency remains. While these buildings produce the effects that Sloterdijk calls for, they do so in a way that is largely out of the control of the inhabitants, uncalibrated. Now then, where's that hand for kneading?